From mboxrd@z Thu Jan 1 00:00:00 1970 From: Claude Code Review Bot To: dri-devel-reviews@example.com Subject: Claude review: Revert "drm/msm/dpu: try reserving the DSPP-less LM first" Date: Sat, 14 Feb 2026 13:59:51 +1000 Message-ID: In-Reply-To: <20260214-revert-dspp-less-v1-1-be0d636a2a6e@oss.qualcomm.com> References: <20260214-revert-dspp-less-v1-1-be0d636a2a6e@oss.qualcomm.com> X-Mailer: Claude Code Patch Reviewer Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Overall Series Review Subject: Revert "drm/msm/dpu: try reserving the DSPP-less LM first" Author: Dmitry Baryshkov Patches: 1 Reviewed: 2026-02-14T13:59:51.835481 --- This is a single-patch revert of commit 42f62cd79578 ("drm/msm/dpu: try reser= ving the DSPP-less LM first"), which introduced an optimization to prefer lay= er mixers (LMs) without DSPPs when DSPPs weren't needed by the topology. The = revert is motivated by reports of blue screens on Hamoa (X1E80100) laptops, i= ndicating that the optimization exposed a hardware dependency or initializati= on issue with higher-numbered LMs on newer DPU generations. The revert is clean and straightforward. It removes the `dpu_rm_find_lms()` h= elper function and inlines the mixer search loop back into `_dpu_rm_reserve_l= ms()`, removing the `skip_dspp` parameter and the two-pass search strategy. T= he resulting code matches what the pre-optimization code looked like, restori= ng the simple linear scan of mixer blocks. No significant issues found. The patch is a correct revert addressing a real = regression. --- Generated by Claude Code Patch Reviewer